Patients with type 1 diabetes (T1DM) are exposed to the risk of experiencing severe episodes of hypoglycemia (HG) as they undergo insulin treatment to maintain the glucose control that is necessary to prevent diabetes complications. Recurrent HG can lead to development of the hypoglycemia-associated autonomic failure (HAAF) and to the syndrome of HG unawareness, where patients lose the ability to recognize symptoms of HG until they become unable to treat themselves. Strict avoidance of HG partially restores awareness of HG, however it is very difficult to achieve. Drugs and therapies that directly target the brain mechanisms responsible for HG unawareness have promise as better ways to prevent and/or reverse HAAF, thus crucially advancing care for patients with diabetes. In order to effectively design and monitor such treatments, it is necessary however to first characterize the physiological and pathological brain responses to HG, a research topic that has been only partly covered by current literature and that is at the center of the current proposal. I this project we will utilize state-of-the-art MRI methods to monitor non-invasively the brain responses during controlled experimental HG in a subject population which includes HG-unaware and HG-aware T1DM subjects, and healthy controls with and without a treatment that induces HAAF. Hypoglycemia-induced brain activations will be identified during different phases of HG by changes in cerebral blood flow. Multiple brain functionally-connected networks will also be monitored by using functional MRI based on the blood oxygenation level contrast, with a novel imaging protocol recently developed within the Human Connectome Project. Our central hypothesis is that altered brain responses to HG in HG-unaware T1DM subjects, as indentified by measures of brain activation patterns and functionally- connected brain networks, are a result of antecedent HG episodes, not diabetes. Therefore we expect that the activation patterns and the brain networks of HG-unaware T1DM subjects will be different from healthy and HG-aware T1DM subjects, but will resemble those of healthy controls who undergo HAAF induction. Our research team is uniquely positioned to conduct this research, as it has gained extensive expertise in mastering high magnetic field MR applications, and holds a long track record of MR research in diabetes. Ultimately, the study design of this proposal will allow identifying with unprecedented sensitivity the altered brain responses to HG in those T1DM patients who are unable to recognize the symptoms of HG, thus filling a critical gap of knowledge which will have significant impact in defining strategies that will reduce the complications of management of diabetes. Since our study has been designed to be conducted on a clinical 3 Tesla scanner, the measurements that we utilize here to detect the brain responses to HG can be easily extended to future clinical trials aiming to monitor the efficacy of new treatments targeting the brain substrates of HG.